Enediyne antibiotics, which damage nucleic acids by radical-based mechanisms, are among the most potent antitumor agents known and, as conjugates of monoclonal antibodies, show promise in ongoing phase III clinical trials in patients with acute myelogenous leukemia. Mechanistic studies have uncovered novel types of nucleic acid damage and ways of controlling cell growth, viral replication, gene expression, and apoptosis. The recent discovery that base-catalyzed activation of the enediyne neocarzinostatin chromophore to a unique wedge-shaped ('double-decker') radical species that specifically binds and cleaves nucleic acid bulges, a conformational form involved in a wide range of important biological processes, and the elucidation of the solution structure of the complex formed between its analog and a two-base DNA bulge provide insights into the design of bulge-specific molecules. This research will focus on the design and synthesis of molecules, including ones with functional moieties for covalent interactions, specific for various bulges in nucleic acids, including that of TAR RNA of HIV in AIDS. Bulged structures, implicated in slipped replication of unstable nucleotide repeats in neurodegenerative diseases and cancers, will also be explored as targets for these agents. Biological systems will be developed to study the ability of bulge-specific molecules to block the formation of toxic proteins containing glutamine tracts due to expansion of the' triplet CAG, as occurs in Huntington's disease. Computer-based modeling, NMR spectroscopy, and screening of RNA diversity libraries for ligand-binding aptamers will be used to enhance molecular specificity. Enediynes will be studied as probes of other unusual structures, such as single-nucleotide bulges (implicated in frame- shift mutagenesis), bubbles, triplexes, etc. Enediynes also form novel covalent monoadducts and interstrand cross-links with nucleic acid sugars. The precise chemistry of adduction and the biochemical mechanisms involved in cellular repair will be explored. Evidence that adducts on DNA deoxyribose are not recognized by the cellular nucleotide excision repair system may explain the high drug toxicity and be the basis of a new type of cancer chemotherapy.